Entrepreneurs play a pivotal role in enhancing productivity, innovation and job creation. In my recent research with Andrea Sy (CEMFI, Madrid), we explore financial constraints on new businesses and, in particular, gender disparities in US firms’ access to credit and gender-driven capital misallocation (Morazzoni and Sy, 2022).
We analyse data from the Kauffman Firm Survey, which follows 4,928 American businesses founded in 2004 through their first eight years. The survey contains detailed information about the owners, company accounts, sources of finance, and outcomes of loan applications. This allows us to compare women- and men-led businesses with similar owners, size, industry, location and legal structure.
Who applies for business loans – and who gets them?
Women are no less likely to apply for a business loan. But among applicants, 32% of women-led firms are rejected, compared with 19% of men-led firms. After accounting for differences such as the owner’s education and experience, company size, existing debt and credit score, women remain about 10 percentage points more likely to be turned down.
This difference is difficult to explain through business risk. In our sample, women-led firms have better credit risk scores and are not more volatile. They also have higher profit margins. Yet they carry less business debt, without making up the difference through additional equity finance.
Limited borrowing appears to affect how much capital these companies use. By capital, we mean productive assets such as equipment and machinery. Women-led firms generate between 8% and 12% more revenue from each dollar of capital than comparable firms led by men. This sounds positive, but it is also a warning sign: an extra dollar invested in these firms could generate a relatively high return, suggesting that they are operating with too little capital.
The pattern is not simply about women choosing different industries. It remains when we compare firms within detailed sectors. It is also weaker in US states where women make up a larger share of business owners: women-led firms in these parts of the country hold more debt and the difference in returns to capital is smaller. Together, these findings connect unequal access to finance with an inefficient distribution of business investment.
Why does this matter beyond individual firms?
To measure the wider cost, we build a model in which people differ in wealth and business ability, and decide whether to become workers or entrepreneurs. The key difference between women and men in our baseline analysis is how much they can borrow against their own assets. We then match the model to evidence from US firms and households.
The model reproduces most of the observed gender difference in firms’ use of capital, although it explains only about one-third of the gap in business ownership. That limitation is informative: unequal credit access appears to be important, but it is not the only barrier that women face when starting or running companies.
We then ask what would happen if equally placed women and men could borrow on the same terms. In this simulation, the number of women entrepreneurs rises by about 9%, while the inefficient distribution of capital falls by about 12%. Total output increases by 3.8%, although this is a modelled ‘what-if’, not a policy forecast.
These estimates are best read as evidence that the gender credit gap may have meaningful economy-wide costs. The model places all production in the entrepreneurial sector, which tends to magnify the result. Adding large companies that are less dependent on owners’ personal borrowing reduces the gain, although it remains sizeable. The benefits also spread beyond women business owners. Better-funded firms demand more workers and capital, raising wages and returns to saving in the model. Overall welfare rises for both women and men, even though some less productive entrepreneurs who are men leave business ownership as competition for resources increases.
What should policy and research do next?
Our policy simulations show why design matters. A broad subsidy to business profits can encourage entry of more men than women because the original borrowing gap remains. Government-backed credit helps women’s entrepreneurship relatively more because it directly eases the financing constraint. Support based only on collateral can favour men if they are still allowed to borrow more against the same assets.
These comparisons are not a ranking of real programmes: they simplify taxes, lenders and administrative costs. But they do suggest that general support for entrepreneurs will not automatically close a gender gap. Policies must address the obstacle that is actually limiting investment.
The biggest unanswered question is why women face tighter credit access. The evidence is consistent with several possibilities, including discrimination, weaker business networks, differences in information or collateral, and wider social norms. It does not establish which cause dominates.
More data, detailed records of lenders’ decisions and credible policy experiments are needed, especially beyond the young firms observed for the period from 2004 to 2011. The broader lesson is already clear: when finance does not reach productive businesses, the cost is borne not only by their owners, but also by workers, consumers and the economy as a whole.
